Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

Annotations only say readOnly=false and destructive=false; the description goes further by disclosing the staging model — bad or blank amounts are held and reported rather than booked as $0, and nothing affects the tenant's numbers until a person certifies the mapping. That is exactly the kind of consequence an agent needs before calling. It stops short of covering re-ingest/duplicate behavior or partial-failure semantics.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Is the description appropriately sized, front-loaded, and free of redundancy?

Two dense but well-ordered sentences: the action and file taxonomy come first, then the operational guarantees. Every clause carries information (validation behavior, human certification, account requirement), with minimal filler.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

For a 4-parameter mutating tool with a nested mapping object and no output schema, the description covers the account prerequisite, the validation/staging model, and human certification. What an agent gets back from a held-rows report, and whether re-ingesting the same file duplicates data, are the remaining gaps.

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

Schema description coverage is 100% and the description largely restates what the schema already documents (the three kinds, mapping your columns to Aurgus fields). It adds the useful framing that the mapping is what a human later certifies, but no new syntax or validation rules beyond the schema.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Does the description clearly state what the tool does and how it differs from similar tools?

States a concrete verb ('hand ... to read into your tenant') and enumerates the three accepted file kinds with definitions for each. It is clearly an ingestion tool for rebate extracts, though it never distinguishes itself from the sibling ingest_agreement, so an agent must infer the boundary.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Does the description explain when to use this tool, when not to, or what alternatives exist?

It tells the agent what to supply (raw CSV, kind, column mapping) and gives a hard prerequisite ('Requires a connected Aurgus account'), which is useful gating. But there is no when-not-to-use guidance and no routing versus ingest_agreement or receive_document, leaving the choice between ingestion siblings to inference.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.